Hi,
We have a processor which reads all our wavefiles, cross references it with a spreadsheet containing all the text for the wavefiles and outputs a textfile containing the text in the wavefile so it can be processed by the FaceFx tool.
ie.
order_move.wav generates order_move.txt in the same location as the wavefile
Is there any way for us to add tags in the text file to denote that the audio is shouting, or is this handled by FaceFx parsing particular characters such as exclamation marks in the text file? We have a naming convention for our wavefiles that denotes "shouted" audio that we can use to add this tag.
Many thanks
Check out these documentation pages on text tags:
Emoticons
Event Text Tags
Curve Text Tags
FaceFX automatically generates events with phrase-ending punctuation (! , . ?). The event has the duration of the phrase. So try analyzing audio with text that has an exclamation point, then going to the event tab and double-clicking on the PhraseEvents|! event to create the associated animation. Add a curve to that animation that is about one second long so that when it is multiplied by the duration of the phrase it will be correct.
Another way to go about it would be to use an analysis text preprocessor to look at the name of the audio or other triggers in the text to insert tags into the text automatically.